def upgrade_languages(root, data):
langs = XPath('./opf:metadata/dc:language')(root)
if langs:
for lang in langs:
lang.attrib.clear()
else:
# EPUB spec says dc:language is required
metadata = XPath('./opf:metadata')(root)[0]
l = metadata.makeelement(DC('language'))
l.text = 'und'
metadata.append(l)
def upgrade_series(root, data):
series, series_index = None, '1.0'
for meta in XPath('./opf:metadata/opf:meta[@name="calibre:series"]')(root):
remove_element(meta, data.refines)
series = meta.get('content')
for meta in XPath('./opf:metadata/opf:meta[@name="calibre:series_index"]')(
root):
remove_element(meta, data.refines)
series_index = meta.get('content')
if series:
create_series(root, data.refines, series, series_index)
def upgrade_cover(root, data):
for item in XPath('./opf:metadata/opf:meta[@name="cover"]')(root):
remove_element(item, data.refines)
item_id = item.get('content')
for item in XPath(
'./opf:manifest/opf:item[@id and @href and @media-type]')(
root):
if item.get('id') == item_id:
mt = (item.get('media-type') or '').lower()
if mt and 'xml' not in mt and 'html' not in mt:
item.set(
'properties',
normalize_whitespace((item.get('properties') or '') +
' cover-image'))
def upgrade_authors(root, data):
for which in 'creator', 'contributor':
for elem in XPath('./opf:metadata/dc:' + which)(root):
role = elem.attrib.pop(OPF('role'), None)
sort = elem.attrib.pop(OPF('file-as'), None)
if role or sort:
aid = ensure_id(elem)
metadata = elem.getparent()
if role:
m = metadata.makeelement(OPF('meta'),
attrib={
'refines': '#' + aid,
'property': 'role',
'scheme': 'marc:relators'
})
m.text = role
metadata.append(m)
if sort:
m = metadata.makeelement(OPF('meta'),
attrib={
'refines': '#' + aid,
'property': 'file-as'
})
m.text = sort
metadata.append(m)
def upgrade_rating(root, data):
rating = None
for meta in XPath('./opf:metadata/opf:meta[@name="calibre:rating"]')(root):
remove_element(meta, data.refines)
rating = meta.get('content')
if rating is not None:
create_rating(root, data.prefixes, rating)
def upgrade_cover(root, data):
for item in XPath('./opf:metadata/opf:meta[@name="cover"]')(root):
# Google Play Books does not recognize covers unless the old style
# <meta name="cover"> is present, so leave it in
# remove_element(item, data.refines)
item_id = item.get('content')
for item in XPath(
'./opf:manifest/opf:item[@id and @href and @media-type]')(
root):
if item.get('id') == item_id:
mt = (item.get('media-type') or '').lower()
if mt and 'xml' not in mt and 'html' not in mt:
item.set(
'properties',
normalize_whitespace((item.get('properties') or '') +
' cover-image'))
def upgrade_date(root, data):
found = False
for date in XPath('./opf:metadata/dc:date')(root):
val = date.text
if val:
found = True
continue
if not val or found: # only one dc:date allowed
remove_element(date, data.refines)
def upgrade_title(root, data):
first_title = None
for title in XPath('./opf:metadata/dc:title')(root):
if not title.text or not title.text.strip():
remove_element(title, data.refines)
continue
if first_title is None:
first_title = title
title_sort = None
for m in XPath('./opf:metadata/opf:meta[@name="calibre:title_sort"]')(root):
ans = m.get('content')
if ans:
title_sort = ans
remove_element(m, data.refines)
if first_title is not None:
ts = [refdef('file-as', title_sort)] if title_sort else ()
set_refines(first_title, data.refines, refdef('title-type', 'main'), *ts)
def upgrade_timestamp(root, data):
for meta in XPath('./opf:metadata/opf:meta[@name="calibre:timestamp"]')(root):
m = meta.getparent()
remove_element(meta, data.refines)
val = meta.get('content')
if val:
try:
val = parse_date(val, is_w3cdtf=True)
except Exception:
pass
else:
create_timestamp(root, data.prefixes, m, val)
def upgrade_identifiers(root, data):
for ident in XPath('./opf:metadata/dc:identifier')(root):
val = (ident.text or '').strip()
lval = val.lower()
scheme = ident.attrib.pop(OPF('scheme'), None)
if lval.startswith('urn:'):
prefix, rest = val[4:].partition(':')[::2]
if prefix and rest:
scheme, val = prefix, rest
if scheme and val:
ident.text = '{}:{}'.format(scheme, val)
for attr in tuple(ident.attrib):
if attr != 'id':
del ident.attrib[attr]
def upgrade_languages(root, data):
langs = XPath('./opf:metadata/dc:language')(root)
if langs:
for lang in langs:
lang.attrib.clear()
else:
# EPUB spec says dc:language is required
metadata = XPath('./opf:metadata')(root)[0]
l = metadata.makeelement(DC('language'))
l.text = 'und'
metadata.append(l)
def upgrade_meta(root, data):
for meta in XPath('./opf:metadata/opf:meta[@name]')(root):
name, content = meta.get('name'), meta.get('content') or ''
if name.startswith('rendition:'):
name = name.partition(':')[-1]
prop = None
if name in ('orientation', 'layout', 'spread'):
prop = 'rendition:' + name
elif name == 'fixed-layout':
prop = 'rendition:layout'
content = {'true': 'pre-paginated'}.get(content.lower(), 'reflowable')
elif name == 'orientation-lock':
prop = 'rendition:orientation'
content = {'portrait': 'portrait', 'landscape': 'landscape'}.get(content.lower(), 'auto')
if prop:
del meta.attrib['name']
del meta.attrib['content']
meta.set('property', prop)
meta.text = content
def test_identifiers(self): # {{{
def idt(val, scheme=None, iid=''):
return '<dc:identifier id="{id}" {scheme}>{val}</dc:identifier>'.format(scheme=('opf:scheme="%s"'%scheme if scheme else ''), val=val, id=iid)
def ri(root):
return dict(read_identifiers(root, read_prefixes(root), default_refines))
for m, result in (
(idt('abc', 'ISBN'), {}),
(idt('isbn:9780230739581'), {'isbn':['9780230739581']}),
(idt('urn:isbn:9780230739581'), {'isbn':['9780230739581']}),
(idt('9780230739581', 'ISBN'), {'isbn':['9780230739581']}),
(idt('isbn:9780230739581', 'ISBN'), {'isbn':['9780230739581']}),
(idt('key:val'), {'key':['val']}),
(idt('url:http://x'), {'url':['http://x']}),
(idt('a:1')+idt('a:2'), {'a':['1', '2']}),
):
self.ae(result, ri(self.get_opf(m)))
root = self.get_opf(metadata=idt('a:1')+idt('a:2')+idt('calibre:x')+idt('uuid:y'))
mi = read_metadata(root)
self.ae(mi.application_id, 'x')
set_application_id(root, {}, default_refines, 'y')
mi = read_metadata(root)
self.ae(mi.application_id, 'y')
root = self.get_opf(metadata=idt('i:1', iid='uid') + idt('r:1') + idt('o:1'))
set_identifiers(root, read_prefixes(root), default_refines, {'i':'2', 'o':'2'})
self.ae({'i':['2', '1'], 'r':['1'], 'o':['2']}, ri(root))
self.ae(1, len(XPath('//dc:identifier[@id="uid"]')(root)))
root = self.get_opf(metadata=idt('i:1', iid='uid') + idt('r:1') + idt('o:1'))
set_identifiers(root, read_prefixes(root), default_refines, {'i':'2', 'o':'2'}, force_identifiers=True)
self.ae({'i':['2', '1'], 'o':['2']}, ri(root))
root = self.get_opf(metadata=idt('i:1', iid='uid') + idt('r:1') + idt('o:1'))
set_application_id(root, {}, default_refines, 'y')
mi = read_metadata(root)
self.ae(mi.application_id, 'y')
def remove_invalid_attrs_in_dc_metadata(root, data):
for tag in XPath('//*[namespace-uri() = "{}"]'.format(DC('')[1:-1]))(root):
for k in tuple(tag.attrib):
if k != 'id':
del tag.attrib[k]
